Community site CMS - what features?

In a perfect world, where MS gave Sun 2,000 engineers to make a new version of JGO (i.e. this site), what features would you want to see? (e.g. look at JGF’s current features).

I’m having a hard time working out a short-list of CMS’s to use for the next generation of JGF. The aim of moving to a CMS (Content Managament System) is so that you can have more involvement, and so that various cool features already proposed by people can be added.

So…what do you want?

Submit/Update game entries automagically. I.e. through some sort of moderation process but just with online forms.

Submit/Update articles automagically. Much like above.

Code repository for clever snipets.

News submittal.

Approval panel selected from JGO folks with a voting process and a weeding out of infrequent visitors.

Server API, highscore and usage reporting system that can be maintained in one central place. Provide a Java API to access it remotely from games.

Search/Indexing of the existing materials.

NO FORUMS! :slight_smile:

Kev

submitHostedGameServer( serverjar )

[quote]submitHostedGameServer( serverjar )
[/quote]
Oh yeah, the porn industry would love that! Free webservices on a high-bandwidth site. :-X

For j3d.org, I just use CVS with a couple of custom scripts to keep the various parts of the site up to date. However, that requires some form of manual updates from the maintainers.

About the only thing I would have liked is an automated system for submitting news and site links from a web form. Also everything else needs vetting in some form or another.

Perhaps one thing that may be useful is to go with Slashcode or something similar. That way the articles/links/news etc can each have their own page, but users can provide comment directly on it. One of the things I find most useful when reading other sites is the ability to read the review and then see the user comments directly under it. Sometimes they are completely at odds with the posted review/article and can provide extra insight, particularly with regards to bias.

How about convert the site to use JSP / Servlets ?

That would allow the community to help build in features, etc.

I like Kev’s ideas above too!

  1. Hosted game servers would probably have to be authorised on a source-code basis, and would probably have to be compatible with the grexengine containers.

However, it might be possible to write a generic simple fast server like Herk’s stuff (or even, depending on his interest, actually use Herk’s stuff :)) and functioned somewhat like an “online HashMap” - i.e. developers/game-authors get to put data in, and pull it out, but not to launch executable code on the server. This way, security concerns are irrelevant (modulo suitable hardening of the server side), and as long as authors didn’t mind doing their logic client-side they’d get both a free server AND avoid having to write all the networking code. Herk, if you have any thoughts on this, email me. I don’t know enough about your stuff to know how easily it would function as a data-passing-only system.

  1. Kev, when you say “NO FORUMS”, I take it you are speaking in CMSese? I.e. CMS’s tend to have two types of forum (usually the same engine but different parts of the site). One is a traditional free-for-all forum like this, the other is a “commentary”/“discussion” where the site maintainers can embed a mini-forum in any page, purely to discuss what’s on that page.

I thought one feature people wanted was to be able to comment on and review games posted on the site, so these “embedded mini-forums” would seem like a wishlist feature?

Sadly, it seems the free CMS’s today are still mostly written by trendy bloggers who know little about CMS’s in the field, other than adding fancy features. So they haven’t got their heads round really basic stuff like the importance of fine-grained access control - and it seems none of the really nice ones would allow mini forums on each page with a different unique named user as admin for each different mini forum. This is trivial stuff, and it’s deeply depressing that so many CMS’s don’t even “get” this generic concept :(. (although there are some that do, they seem to have other problems instead :().

Leaning more and more towards building a new one from scratch…Yuk.

  1. AFAICS SlashCode etc is far too simplistic / narrow focus, and would provide nowhere near enough features :(. And there appears to be only one CMS based on servlets. Worse, it’s extremely hard to find ISP’s that will host servlets (compare to the number that will host LAMP!) and they have no inherent benefit AFAICS, at least for this. If we have to have a JVM, there are much much better things to be running than servlets ;)…

Interesting point. I’d not expected more than 2, perhaps 3, people to ever devote enough time to actually code for a java gaming site. So far, for instance, we’ve had about 10 serious volunteers for article writing, about 5 for article editing, about 5 finding and sending in other people’s games, and … 3 pointing out site errors. Number volunteering to help fix my broken XSLT: 0.

Which is roughly what I expect. But perhaps if the site were a lot more dynamic, and had a full publishing system (so that anyone could easily make online changes, modulo permissions) this would change, and more people would want to provide code. Do you think so?

What server side language would the CMS be coded in? I’d be willing to set aside a large proportion of my time to help develop the code around it if we are creating it from scratch. I’d also like to help on the designs and such (XHTML + CSS…blah blah or XSLT).